Pythian solves inventory optimization and supply chain issues for a major supermarket chain

What we did

• Created the infrastructure necessary to complete a semi-automated, data-consistent set of predictive model testing against the client’s real, historical data; this involved building a data replenishment pipeline running on Google Cloud (GC) AI Platform and orchestrated by Cloud Composer service along with Apache Airflow and custom Python scripts 
• Built a contingency pipeline running on AI Platform and Cloud Composer 
• Built a machine learning model in TensorFlow 
• Worked collaboratively with the client team to evaluate four predictive models against the company’s historical data 
• Implemented the chosen model to run daily predictions against the client’s sales data 
   

Technologies used